Obesity in preschool children is a well-documented public health crisis, and children's food consumption likely contributes to their excess weight. In previous research, we have shown that preschool children are easily influenced by food advertising on television. However, there has been no attention given to the effects of the food messages in children's toys or classroom activities on their food choices. Many children's toys, including books, crafts, games, academic activities such as counting or coloring pages, imaginary play activities and songs incorporate food messages and most of these messages promote the consumption of sweets. We propose to conduct an experiment to examine the effects of healthful and unhealthful food-related toys and classroom activities on preschool children's choice of either fruit or candy. Secondary outcomes for this experiment will include parent reports of children's requests to purchase foods, and observations of children's consumption of fruit and vegetables in their classroom. The target group for these interventions will be low-income, predominantly Mexican-American children enrolled in Head Start preschool programs. The long-range goal of this research is to develop a classroom environment component, based on the results of this pilot research, to incorporate into our current family-based intervention for preschool children and evaluate this combined intervention in a future randomized controlled trial. The primary hypothesis that will be tested in this experiment is: from a selection of fruit and candy, children will choose more candy while exposed to the unhealthful food toy condition and more fruit while exposed to the healthful food toy condition compared to the no food toy control condition. A sample of 72 children between the ages of 2 years 9 months and 5 years will be recruited from four Head Start classrooms in predominantly Mexican-American neighborhoods. An interview with teachers and an inventory of all food toys in each classroom will be conducted prior to the start of the interventions. A within-subjects, crossover design will be used to compare the two treatment conditions, healthful food toys and activities or unhealthful food toys and activities, to a no food toy control condition. The treatment conditions will be randomly assigned. The control condition, no food toys, will implemented two weeks prior to each treatment condition. Outcome data, including children's choice of four items from a selection of fruit and candy, parents'reports of children's food purchase requests during the past week, and observations of children's fruit and vegetable consumption during classroom meals, will be collected during the last 2 days of each treatment condition. PUBLIC HEALTH RELEVANCE: Children's toys have been recognized as a source of food marketing, yet children's responses to both the healthful and unhealthful food messages in their toys have not been empirically examined, but may impact the effectiveness of school based nutrition interventions. Based on the results of this research, school policies that eliminate toys with unhealthful food messages, promote toys with healthful food messages or both may be crafted. These policies would be relatively simple to implement, require very little staff training and are economically feasible and sustainable.